Entry Name: "UVA-Migut-MC2"

VAST Challenge 2016
Mini-Challenge 2

Team Members:

Bas Chatel, University of Amsterdam, bastiaan.chatel@gmail.com

Lesli Dao, University of Amsterdam, l.dao@hotmail.com

Wessel de Jong, University of Amsterdam, wes_dejong9@hotmail.com

Gosia Migut, University of Amsterdam, mmigut@gmail.com

Student Team: Yes

Tools Used:

Visualizations in D3 developed by the undergraduate student team for the course Minor Programming: Programming Project at the University of Amsterdam.

Adobe Illustrator

Approximately how many hours were spent working on this submission in total?

480

May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2016 is complete? Yes

Video

https://staff.fnwi.uva.nl/m.a.migut/UVA-Migut-MC2.wmv

Questions

MC2.1 – What are the typical patterns visible in the prox card data? What does a typical day look like for GAStech employees? Limit your response to no more than 6 images and 500 words.

The prox data is fairly regular. The detections follow daily patterns and on the weekends that are very few to none detections. On the Saturday 4^th of June there was not a single detection for the whole building. We describe the observed patterns for the general data, for each floor and separately for the weekends.

Figure MC2.1_1. In the big view a daily pattern (3 a.m. – 3 a.m.) of the amount of detections for the three floors; in the overview below the pattern over the two weeks period.

· Figure MC2.1_1 shows the daily pattern of the amount of detections for the three floors. There are always the most people present on the second floor, the least on the third floor. There are people leaving the first floor at 9 a.m. and just before 3 p.m., but it seems they are not going to other floors (not more people detected on other floors at those time stamps).

· The 1^st floor has three peaks during the day, see figure 2. The employees start coming in around 5:30 a.m. with a peak at 7 a.m. Then around 9 a.m. most of them are gone from the floor 1, followed by increased amount of detection with a peak at 12 p.m. Then again they leave the floor, with the lowest amount of detections around 2:15 p.m. followed by increase with a peak just before 4p.m. Afterwards people start leaving the 1^st floor until 6 p.m. From 6 p.m. there are still several detection until everybody is gone at around 12:30 a.m.

The 1^st floor has most detections in the zones: 1and 4, followed by zone 4. Detection in zone 1 and 4 follow the general daily pattern of the 1^st floor. Detections in zone 2 follow different pattern. People start coming in at around 9 a.m. with a peak at 12 p.m. and they all leave by 1:30 p.m., with a very few detections during the rest of the day. There are no detections in zone5, besides on the 31^st of May just before the midday. There are very few detections in zone 7 and most of them after the midday.

Figure MC2.1_2. Detections on the 1^st floor per zone. In the big view a daily patterns, in the overview beneath the overall pattern over the two weeks period.

· The 2^nd floor has a very regular daily patter with most detections in the zones: 1, 4, 6, 7. In the zone 5 there are no detections. In zone 2 there are twice as many detections as in zone 3;

Figure MC2.1_3

· The 3^rd floor has the least regular daily pattern. In general the most detections are in zone 1, 4. In the server room there are slightly more detections on the 10^th of June. In zone 5 there are no detections. Zone 2, 3 and 6 follow the same patterns, only the values for zone 3 are higher than for zone 6. At the beginning of the 2^nd of June zone 1 and 4 have more detections, than normally at the beginning of the day. Also on the 7^th and 8^th of June the number of detections fluctuates differently than on the other days.

Figure MC2.1_4

· Weekends: there are four people (Mbramar, Llagos, Ostrum001, Lcarrara001) that have been detected during the weekend in the building, but the zones they visited were their regular working places during the week days.

Figure MC 2.1_5

· Most employees have fairly regular behavior, visiting the same zones every day. Some have exactly the same amount of detections daily being in at the specific zone and floor. We observe that some employees behave differently, including having more than one prox card or visiting unusual zones, which we describe in MC2.4.

Figure MC2.1_6

MC2.2 – Describe up to ten of the most interesting patterns you observe in the building data. Describe what is notable about the pattern and explain what you can about the significance of the pattern.

Limit your response to no more than 10 images and 1000 words.

1. Floor 2 zone 12a/b/c (corridors) Lights Power constantly on, measurements are high during the day and 0 in the weekend. Zone 2 and 6 the highest values, followed by zone 4 and 9, followed by zones 10, 14, 16. Equipment Power follows the same pattern.

Figure MC2.2_1

Figure MC2.2_2

2. The general data has a lot of stable sensor readings: sensor readings are constant during the whole period for: Cool Schedule Value, Loop Temp Schedule, Heat Schedule Value, Pump Power, Supply Side Inlet Mass Flow Rate, Water Heater Setpoint, Supply Side Outlet Temperature, Water Heater Tank Temperature, Loop Temp Schedule. Interestingly the Water Heater Setpoint and Loop temp Schedule have the same value of 60. The Deli-Fan Power, Drybulb Temp, Supply Side Inlet temp, Supply Side Outlet Temp, Water Heater Tank Temp, Water Heater Gas Rate have regular patterns for the weekdays and for the weekends (weekday vs weekend may very).

3. On the 1^st floor the Lights Power, that is constant over de period in zones 3 (main entrance), 8a/b (corridor) and in the weekends. Also the Equipment Power is constant in the same zones, while the other zones fluctuate heavily, being flat in the weekend.

4. We see an interesting pattern for the bathroom exhaust fan on Sundays (the pattern is regular on the week days). On the 1^st and the 2^nd floor on Sundays the fan is using power the whole day long. On the 3^rd floor the fan is off the whole day on both Sundays.

Figure MC2.2_3 Fan power on the 1^st floor.

Figure MC2.2_4 Fan power on the 3^rd floor.

5. The Night cycles are very regular, see figures MC2.2_Figure MC2.2_5

Figure MC2.2_6

Figure MC2.2_7

MC2.3 – Describe up to ten notable anomalies or unusual events you see in the data. Describe when and where the event or anomaly occurs and describe why it is notable. If you have more than ten anomalies to report, prioritize those anomalies that are most likely to represent a danger or serious issue for building operation. Limit your response to no more than 10 images and 1000 words.

· On the 7^th and 8^th of June we observe anomalies in many sensor readings in the general data and data on all the floors, see figures MC2.3_1 – MC2.3_7:

o For the general data: The Electric Demand Power and The Total Electric Demand show a high peak at 7 a.m. on both days ranging approximately from 50.000 to 250.000 W and 120.000 to 350.000 W respectively (comparing to regular patters on other days not higher then approximately 70.000 W); followed by the deviating values during the weekend: higher measurement values (not higher then approximately 60.000 W) comparing to the previous weekend (not higher then approximately 130.000 W), see figure MC2.3_1 and MC2.3_2.

o On the 1^st floor many sensor readings have deviating values. The Reheat Coil Power shows two high picks during those days (comparing to fairly regular patters on the other days). Outdoor Air Mass Flow Rate does not fluctuate on those two days. Cooling COIL Power is constant at the beginning of these two days. Air Loop Inlet Temp has two peaks on those days. Air Loop Inlet Mass Flow Rate does not fluctuate on those days. Also CO2 Concentrations are have two distinct peaks. There are more variables influenced, however we did not find the relation between those.

o On the 2^nd floor we for example observe anomalies in zone 8 Thermostat Temp, peaks on 2^nd and 8^th of June; op 7^th and 8^th June constant higher than average.

o On the 3^rd floor one example anomaly is Thermostat Temp with two high peaks, and Fan Power with low measurements.

Figure MC2.3_1

Figure MC2.3_2

Figure MC2.3_3

Figure MC2.3_4

Figure MC2.3_5

Figure MC2.3_6

Figure MC2.3_7

· The anomalies on the 7^th and 8^th of June seem to have influence on several sensor readings on the proceeding weekend (as the values differ from the preceding weekend).

On the 1^st floor the second weekend values are higher than the first weekend values for the several sensor reading for example: VAV_SYS COOLING COIL Power, VAV_SYS AIR LOOP INLET Mass Flow Rate, etc. See figure MC2.3_8.

Figure MC2.3_8

· On the 31^st of May we observe anomalies on the 1^st floor in the zone 2: The Lights Power and the Equipment Power showing a pick in the middle of the day (while regularly constant around 0 and 400, respectively), see figure MC2.3_9.

Figure MC2.3_9

· In the general data, the Wind Direction fluctuates around 50 degrees each day, but on 5^th, 8^th and 9^th the fluctuations are around 0 and 360 degrees. Also the Wind Speed shows irregularities with the lower values in the beginning and end of the measured period, see figure MC2.3_10.

Figure MC2.3_10

MC2.4 –– Describe up to five observed relationships between the proximity card data and building data elements. If you find a causal relationship (for example, a building event or condition leading to personnel behavior changes or personnel activity leading to building operations changes), describe your discovered cause and effect, the evidence you found to support it, and your level of confidence in your assessment of the relationship. Limit your response to no more than 10 images and 1000 words.

1. We observe a change in the data on the 3^rd floor on the 2^nd of June. From 2^nd of June the Thermostat Temp and Thermostat Heating& Cooling Setpoint in HVAC zone 1 are fluctuating more comparing to other days, while on the 2^nd of June we observed high peaks in zones 1 and 4 around 7-8 a.m. in the prox data.

2. On the 9^th of June the Reheat COIL Power inHVAC zone 8 is fluctuating heavily and at the same time we see less detections in the prox data in zones 1,2,3,4, 6 and no one in the server room.

3. The weekend are also interesting. Four of the employees were detected on the weekends: (1) Lcarrara001, detected on Sunday 5^th of June on the 1^st floor in the zones 1 and 4, on the 2^nd floor in the zone 2 and on the 3^rd floor in the zones 1, 3, 4 and 6; (2) Ostrum001, detected on Saturday 11^th of June on the 1^st floor in the zones 1 and 4, on the 2^nd floor in the zone 4 and on the 3^rd floor in the zones 1, 3, 4 and 6; (3) Llagos, detected on Sunday 12^th of June on the 1^st floor in the zones 1 and 4, on the 2^nd floor in the zone 4 and on the 3^rd floor in the zones 1, 3, 4 and 6; (4) Mbramar, detected on Saturday 11^th of June on the 3^rd floor in the zones 1, 4 and 6. All zone detections are regular work places for the employees detected during the weekend.

In the weekends the Light Power is on only on the 3^rd floor in de HVAC zones 10 (Sunday 5^th of June) and 2, 8 (Saturday, 11^th of June ) and zone 8 (Sunday, 12^th of June). Interestingly HVAC zone 2 doesn’t correspond to any location in the prox data where the employees present in the weekend were detected (it corresponds to zone 2 in the prox data).

4. Most employees have fairly regular behavior, visiting the same zones every day. Some have exactly the same amount of detections daily being in at the specific zone and floor. We observe that some employees behave differently. Few employees seem to have multiple prox cards, like Florez (002 and 003), Unger (001 and 002), Morlun (001 and 002, also her card has different id: first the last name and then the first later of the first name, while others have first latter of the first name and then the last name). Those people have less regular patter as they are detected on different days with different cards. The prox data of employee Morlun is interesting. The prox cards of employee Morlun are both detected on the same days. On Thursday 2^nd of June the card 002 is detected at the 1^st floor in the zones 1, 2, 3 and the card 001 at the 2^nd floor in the zones 1, 4, 6, 7. Also employee Morlun has no detections with the card 001 on the day before, namely on Wednesday 1^st of June at the 2^nd floor (while all other days that card has very regular pattern). Interestingly, on the same day there is difference in the Thermostat Temperature measures on the 2^nd floor (where Morlun usually works).